Results for “ams_dm_scores_vl”

AI-generated from documented ETRM metadata — verify critical details on the linked pages.

Overview

AMS_DM_SCORES_VL is a multi-lingual (ML) view owned by the APPS schema in Oracle E-Business Suite, delivered as part of the AMS – Marketing product family. It exposes data mining scoring runs, which are the records that describe how a predictive model is executed against a target population of customers, prospects, or accounts. In the ETRM 12.2.2 documentation the object carries a VALID status and is classified as a VIEW, with the documented description "Multi-lingual view for data mining scoring runs."

The "_VL" suffix follows the standard Oracle EBS multilingual convention: the view joins a language-independent base table to its translation table and filters the translation rows by the session language using USERENV('LANG'). This means a report or integration querying AMS_DM_SCORES_VL automatically receives the score name and description in the language of the current user session, without the caller having to write language-specific join logic. Within the marketing data mining flow, scoring runs are created from a model, scheduled, executed, and then measured by statistics such as the total records scored and total positive results, all of which are surfaced through this view.

Underlying Base Objects

The view is defined over two documented base objects, both referenced through APPS synonyms:

• AMS_DM_SCORES_ALL_B – the language-independent base table that stores the operational attributes of each scoring run, including identifiers, status, scheduling information, execution counts, and the descriptive flexfield attribute columns.
• AMS_DM_SCORES_ALL_TL – the translation table that stores the user-facing text attributes, principally SCORE_NAME and DESCRIPTION, keyed by SCORE_ID and LANGUAGE.

The join is an equi-join on SCORE_ID, combined with the language predicate T.LANGUAGE = USERENV('LANG'). The B table supplies the ROWID as ROW_ID and essentially all non-translated columns; the TL table contributes only SCORE_NAME and DESCRIPTION to the projection. Because the definition uses synonyms sourced from APPS, the view should be queried from the APPS schema or by a user granted explicit access, in line with standard EBS security practice.

Key Columns

• SCORE_ID – Primary identifier of the scoring run; the join key between the base and translation tables.
• MODEL_ID – Reference to the data mining model against which the scoring run is executed.
• STATUS_CODE / USER_STATUS_ID / STATUS_DATE – Current lifecycle state of the scoring run and the date it entered that state.
• OWNER_USER_ID – The user who owns the scoring run, useful for workload and audit reporting.
• ORG_ID – Operating unit context, enabling multi-org–secured reporting.
• SCHEDULED_DATE / SCHEDULED_TIMEZONE_ID / SCORE_DATE – Scheduling and actual execution timestamps for the run.
• TOTAL_RECORDS / TOTAL_POSITIVES – Volume of records scored and count of positive outcomes, used to assess model lift and response.
• MIN_RECORDS / MAX_RECORDS / ROW_SELECTION_TYPE / EVERY_NTH_ROW / PCT_RANDOM – The population sampling rules applied when selecting rows for scoring.
• RESULTS_FLAG / LOGS_FLAG – Indicators controlling whether result and log artifacts are retained.
• WF_ITEMKEY – Workflow item key linking the run to its Oracle Workflow process instance.
• ATTRIBUTE_CATEGORY through ATTRIBUTE15 – Descriptive flexfield columns for customer-specific extensions.
• SCORE_NAME / DESCRIPTION – Translated, user-facing labels for the scoring run.
• ROW_ID – Exposed ROWID of the base table row, useful for tightly targeted updates or diagnostics.

Common Use Cases and Queries

Reporting on scoring run activity is the dominant use case: monitoring run status, execution volumes, and positive-result rates by model or operating unit. A representative query listing recent scoring runs and their outcome statistics is:

• SELECT score_id, score_name, status_code, score_date, total_records, total_positives FROM apps.ams_dm_scores_vl WHERE org_id = :p_org_id ORDER BY score_date DESC; — a standard status and performance report scoped to an operating unit.
• SELECT model_id, COUNT(*) runs, SUM(total_records) records_scored, SUM(total_positives) positives FROM apps.ams_dm_scores_vl GROUP BY model_id; — model-level effectiveness summary across all completed scoring runs.
• SELECT score_id, score_name FROM apps.ams_dm_scores_vl WHERE status_code = 'SCHEDULED' AND scheduled_date < SYSDATE; — identification of overdue or stalled scheduled runs for operational follow-up.
• SELECT score_id, wf_itemkey, status_code FROM apps.ams_dm_scores_vl WHERE wf_itemkey IS NOT NULL; — correlation of scoring runs with their Oracle Workflow instances for troubleshooting.

Because the view is read-only and multilanguage-aware, it is well suited to concurrent program data sources, OBIEE or BI Publisher datasets, and inbound integration extracts that require localized score names. Updates and inserts must always be directed at the underlying AMS_DM_SCORES_ALL_B and AMS_DM_SCORES_ALL_TL tables, not the view. Queries should filter on ORG_ID where multi-org security applies and should expect the SCORE_NAME and DESCRIPTION values to change with the session language, which is expected behavior rather than a data inconsistency.